INDEX

Explanations

proper nouns and locations

oai_token-act-pair · gpt-3.5-turbo

phrases related to geographical locations and their governance

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

loads

-0.56

ÃĥÃĤ

-0.56

ngth

-0.54

 wisely

-0.54

 alone

-0.54

perhaps

-0.54

 furthermore

-0.54

ãĤ´ãĥ³

-0.54

vertisement

-0.53

wu

-0.53

POSITIVE LOGITS

 same

0.85

 latter

0.81

 aforementioned

0.80

 United

0.76

 nation

0.74

 Philippines

0.74

 country

0.74

 latest

0.74

 city

0.71

 upcoming

0.71

Activations Density 0.799%

proper nouns and locations

phrases related to geographical locations and their governance

No Comments

No Known Activations